The Police National DNA Database: Balancing Crime Detection, Human Rights and Privacy
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
The Police National
DNA Database:
Balancing Crime Detection, Human
Rights and Privacy
A Report by GeneWatch UKThe Police National DNA
Database:
Balancing Crime Detection,
Human Rights and Privacy.
A Report for GeneWatch UK by Kristina Staley
January 2005
The Mill House, Manchester Road, Tideswell, Buxton
Derbyshire, SK17 8LN, UK
Phone: 01287 871898 Fax: 01298 872531
GeneWatch
E-mail: mail@genewatch.org UK
Website: www.genewatch.orgAcknowledgements
GeneWatch would like to thank Jan van Aken, Sarah Sexton and Paul Johnson
for their helpful comments on a draft of this report. Kristina Staley would
also like to thank Val Sales for her help in preparing the report. The content of the final report
remains the responsibility of GeneWatch UK.
Cover photograph
DNA genetic fingerprinting on fingerprint blue backdrop. © Adam Hart-Davis,
http://www.adam-hart-davis.org/
GeneWatch UK
2 January 2005Contents 1 Executive summary ................................................................................................................5 2 Introduction ...........................................................................................................................10 3 What is the National DNA Database?..................................................................................11 3.1 Using DNA to identify individuals ...................................................................................11 3.2 How the police use DNA................................................................................................12 3.3 Concerns about police use of DNA our right to privacy................................................12 4 What information is stored on the NDNAD? How is this information obtained? ...........14 4.1 DNA profiles ..................................................................................................................14 4.2 Obtaining a DNA profile from a tissue sample...............................................................15 5 How is the NDNAD used to identify suspects? How reliable is DNA evidence? ............18 5.1 Using the NDNAD to identify suspects ..........................................................................18 5.2 DNA profiling is not foolproof.........................................................................................19 5.3 Using DNA evidence in court.........................................................................................23 6 What laws govern the use of the NDNAD? What are the limits on police powers? .......25 6.1 England and Wales: legislation relating to the NDNAD .................................................25 6.2 How England and Wales compare to other countries....................................................27 6.3 Concerns about the legislation protecting our civil liberties..........................................28 7 The future of DNA profiling..................................................................................................30 7.1 Predicted changes to DNA profiling...............................................................................30 7.2 Using DNA profiles to predict the characteristics of suspects .......................................31 8 The future of the NDNAD......................................................................................................35 8.1 The changing role of commercial companies ................................................................35 8.2 Links to other national databases..................................................................................35 8.3 Using the NDNAD for other purposes ...........................................................................36 9 Are our human rights and civil liberties being adequately protected? ...........................38 9.1 Whose profiles should be added to the NDNAD? .........................................................38 9.2 When should samples be destroyed? ...........................................................................43 9.3 How should sensitive genetic information be protected ................................................44 9.4 Who decides how the NDNAD should be used? ...........................................................47 10 Conclusions ..........................................................................................................................47 11 References ............................................................................................................................51 List of Boxes Box A: Analysing short tandem repeats (STRs) to generate a DNA profile .............................15 Box B: Use of familial searches to find suspects .....................................................................19 Box C: Poor practice in US police laboratories leads to wrongful convictions .........................20 Box D: Misinterpreting DNA profiles leads to wrongful convictions .........................................22 Box E: A false DNA match leads to wrongful arrest .................................................................23 Box F: The use of the NDNAD in successful police operations ..............................................24 Box G: A brief legal history of the NDNAD ..............................................................................26 Box H: Promoting public support for changes to forensic DNA legislation ...............................27 Box I: New anti-terrorism measures and civil rights................................................................29 Box J: The potential for DNA profiling to reinforce racial prejudice ........................................32 Box K: Using DNA to predict a suspect's origins .....................................................................34 Box L: Government plans to develop national identity registers ..............................................35 GeneWatch UK January 2005 3
Box M The limitations of studying genetic links to criminal behaviour .....................................37
Box N: Taking samples from people who might reoffend .........................................................39
Box O: The Police National Computer......................................................................................40
Box P: Taking DNA samples from peaceful protesters ............................................................41
Box Q: What happens to people after police arrest? ................................................................41
Box R: Freedom to give consent to providing a DNA sample ..................................................42
Box S: Misuse of a sample taken for DNA profiling .................................................................46
Box T: Police refuse to give samples to forensic databases....................................................46
GeneWatch UK
4 January 20051 Executive Summary
Using DNA to trace people who are suspected of committing a crime is one of the biggest
advances in tackling crime since fingerprinting. When DNA profiling is used wisely it can bring
major benefits to society by helping to convict serious criminals including murderers and
rapists. Concerns arise, however, when tissue samples, genetic information and personal data
are stored indefinitely on a DNA database, like the police forensic database known as the
National DNA Database (NDNAD). There are fears that this information may be misused in
ways that threaten our individual rights as well as those of our families. We must be confident
that the police and the Government use DNA in a way that respects our fundamental right to
privacy and protects our civil liberties.
The limits on police powers relating to the use of the NDNAD are enshrined in law. On 4 April
2004, these powers were extended. In England and Wales, the police are now allowed to take
samples without consent from anyone who is simply arrested on suspicion of any recordable
offence. This includes all but the most minor crimes. The new legislation also allows the police
to keep this information indefinitely, even if the person arrested is never charged. This means
the database now contains DNA profiles from 2.1 million individuals and is the most extensive
DNA database in the world. No other police force has greater freedom to obtain, use and store
genetic information from its citizens.
This report reviews the operation of the forensic DNA database in England and Wales, the
benefits it may bring and the concerns that it raises. As the NDNAD continues to expand,
GeneWatch UK believes there is a need for wider public debate on the use of forensic DNA
and makes a number of recommendations for improvements in the retention and handling of
genetic information by the police. We are also concerned about likely changes in the future,
including suggestions that everyone's DNA profile should be added to the database, and the
introduction of new technology that could expand the use of genetic information. GeneWatch
UK hopes this report will encourage and inform society's deliberations on these difficult and
sensitive subjects.
DNA in crime detection
The NDNAD is a database used by the police that contains DNA data held on a computer and
stored samples from the scenes of crimes as well as from people who have been arrested,
acquitted or convicted of a crime. The DNA from every sample is analysed in a limited number
of areas to produce a 'DNA profile' for each individual or sample from the scene of a crime.
The database relies on the fact that every person's DNA is unique (unless they are an identical
twin) and that the chance of an identical match between the DNA sequence taken from two
different people is very low – less than one in a billion if they are not related. However, it is
possible to get 'false' matches and the chances of a 'false' match are increased (i.e. the wrong
person is identified) if samples are obtained from people who are related or if the DNA profile
is incomplete. Incomplete profiles are often obtained from scene of crime samples because the
DNA is often degraded.
Following collection and analysis of a sample, a new DNA profile can be compared with the
profiles stored on the database. A match can arise in four different ways. If:
1. a new scene of crime profile matches the profile of an individual already on the
database. This can help to identify a potential suspect very rapidly;
2. a new individual's profile matches a stored scene of crime profile from an unsolved crime
or crimes on the database. This type of 'speculative search' can identify a potential
suspect long after a crime has been committed;
GeneWatch UK
January 2005 53. a new scene of crime profile is found to match that of an old crime scene. This can help
the police in their investigations even if a suspect is not identified; and
4. the person has been sampled and included on the database previously under the same
or a different name.
Therefore, DNA profiling is a powerful tool in solving many serious crimes and deterring
serious criminals. In a typical month, matches are found linking suspects to 30 murders, 45
rapes and 3,200 motor vehicle, property and drug crimes.
Concerns about the use of DNA profiling and the National DNA Database
There are concerns about the use of this technology since DNA is very different from other
types of forensic data – it has the capacity to reveal a lot more personal information. For
example, DNA data may reveal whether a person is at risk of ill-health as well as who they are
related to. Giving the police or state access to DNA data and samples can therefore give them
much more information about a person than simply who that person is.
The major concerns about the use of the NDNAD relate to:
Þ Balancing the rights of individuals and the interests of society
Society has an interest in reducing crime. Most people want criminals to be caught, detained if
necessary and, if possible, rehabilitated so they do not commit further offences. A temporary
removal of some rights is widely agreed to be a reasonable punishment for committing a
serious crime. But current use of the NDNAD has the potential to threaten the individual's right
to privacy and civil liberties much more widely.
The biggest threat to people's right to privacy is posed by the indefinite retention of DNA
samples, because these could provide unlimited amounts of genetic information about known
individuals. The usefulness of retaining samples after the DNA has been analysed is
questionable, even in the case of convicted criminals. If a person is found to be innocent, there
does not seem to be any compelling reason why their sample should be retained and their
rights to privacy should be denied. The law in England and Wales is unique in allowing
samples from large numbers of innocent people to be retained indefinitely.
The NDNAD also threatens our civil liberties because of its potential to be used as an
instrument of surveillance. Expanding the database puts increasing numbers of people on a
permanent 'list of suspects' even though they may never have been charged or convicted of a
crime. This may subtly alter the way they are viewed both by the state and by their fellow
citizens, potentially undermining the principles of 'innocent until proven guilty' and of
rehabilitation. The NDNAD is also the only database that keeps a person's 'criminal' record
indefinitely, no matter how trivial their offence. This raises the concern that records on the
NDNAD could be used in future to restrict people's rights and freedoms, for example to make it
difficult for them to obtain employment.
Þ The increasing threats to our 'genetic privacy'
The current DNA data used for identification purposes contains very limited information about
a person's genes. However, this may change in the future with plans to use new technology to
exploit the information in DNA samples. Some advocates have argued that this technology will
be able to predict the characteristics of a suspect from the DNA evidence at the scene of a
crime, generating a description along the lines of 'a tall man, with red hair, blue eyes, who's
probably overweight'. Researchers are also looking at predicting ethnicity and health status.
Some even believe it will be possible to predict a person's personality or behaviour. However,
there are serious scientific problems with most of these approaches. Not only is some of the
research fundamentally flawed, much of it is unlikely to produce particularly useful or accurate
predictions. There is also a danger that the information will be used selectively to reinforce
GeneWatch UK
6 January 2005existing prejudices, for example about race or skin colour. Nevertheless, a few genetic tests can reveal important information about some people's health. If use of this new technology were expanded to stored samples from known individuals on the database, the increase in police access to genetic information could pose an even greater threat to privacy. Þ A lack of transparent mechanisms of governance and oversight Some uses of the NDNAD are particularly controversial or sensitive. For example, familial searches can be used to trace suspects if they have any relatives on the database. Such an investigation can represent a major intrusion into family life. There is a risk it may uncover family relationships that people do not know about, including cases of non-paternity. However, as yet there are no guidelines as to when such an approach can be considered ethical and what the implications might be for data protection. This situation is inadequate and open to abuse. Similarly, researchers using the NDNAD do not have to seek consent from participants or the approval of an independent ethics committee to carry out their research. They have only to seek permission from the NDNAD Board. Some of the research could be highly controversial, for example research on ethnicity and criminal behaviour. There is no guarantee that such controversial projects might not be undertaken in the future. The main organisation currently carrying out forensic research, the Forensic Science Service, is also heavily represented on the NDNAD Board. This creates an unacceptable conflict of interest. Þ Increasing police and Government access to personal data Other national databases are being planned and developed, including the National Identity Register to support the use of ID cards, and the new NHS Electronic Care Record Service, which may contain some genetic data in the future. It is not clear under what circumstances the police will be allowed access to this information. Nor is it clear whether any of these databases will be linked, possibly allowing other Government bodies to find out who is on the NDNAD. Expanding and/or linking these databases would give the state unprecedented abilities to monitor the UK population, greatly increasing the threats to our privacy. There are concerns that this access could all too easily be abused, taking the UK closer towards an oppressive 'police state'. Þ Errors in DNA profiling There is no such thing as an error-free database. Mistakes can lead to 'false positives' where an innocent person is wrongly identified. A 'trawl of the database' is not enough to secure a conviction in court: a fresh sample from the accused and corroborating evidence is also needed. But in some cases DNA evidence can be difficult to interpret, particularly when samples from the crime scene are degraded or contain more than one person's DNA. Currently, the criminal justice system may not always take sufficient account of the possibility of errors and people may be wrongly convicted either by mistake or even by being 'framed'. Recommendations –getting the balance right At GeneWatch UK, we recognise the benefits of the NDNAD but we also believe that its current operation does not strike an appropriate balance between the rights of the individual and the interests of the public. While there is no doubt that society does have an interest in the detection and prevention of crime, this cannot be used to justify every infringement of the individual's right to privacy and the loss of our civil liberties. This is especially true for people who are innocent. In this regard, we believe the UK needs to bring its criminal justice legislation more in line with the rest of Europe. There are changes that could be made to the operation of the NDNAD which would protect people's rights and increase public confidence without compromising its role in fighting crime. GeneWatch UK January 2005 7
Recommendation 1: DNA samples (except samples from the scene of a crime) should
not be retained once an investigation is complete. Only DNA profiles and personal data
need to be on the database to find a 'match' for a criminal investigation. Research uses
of the database itself (profiles and personal data) should be restricted to producing
'quality control' statistics on the type of data that has been added and how the data is
being used.
This recommendation removes concerns that samples could be used for purposes other than
identification, such as research into criminal behaviour, without the individual's consent.
Suspects' samples need to be retained for a certain length of time so they can be analysed
and the profiles can be checked. However, destroying the sample once an investigation is
complete does not in any way restrict future searches for matches. All the information that is
needed is stored in the DNA profile held on a computer. Physical samples do not need to be
retained to prevent errors because a fresh sample must be taken anyway before DNA
evidence can be used in court. Although it can be argued that samples may need to be
reanalysed if the technology is updated, in reality upgrading the DNA profiling system used on
the database seems to be both costly and unnecessary.
The right to consent or refuse to take part in research is an important right for individuals and
for society. It is not necessary to use samples or profiles taken without consent to do legitimate
genetic research. It is also questionable whether the NDNAD provides a robust source of data.
Categories within the database such as 'ethnic appearance' are meaningless for scientific
purposes and the DNA profiles and samples will not be representative of either the general or
the 'criminal' population. Genetic research using the database is therefore likely to be
misleading as well as controversial.
Recommendation 2: An independent body should be set up to review all future
applications to access the data and samples for forensic and non-forensic purposes; to
ensure standards are maintained; and to ensure public accountability and transparency.
We are concerned both about the current use of the NDNAD and the potential for the
increased threats to privacy in the future. We think it is essential that the NDNAD is made
more accountable to the public. We therefore believe that a new independent body should be
created that includes lay representation. This body should be made responsible for deciding
when 'familial searching' is allowed, rather than leaving this decision to the police alone.
This body should also review the merits of methods which attempt to predict the characteristics
of individuals from DNA samples left at the scene of a crime. There is a danger that attempts
to predict physical appearance, or other characteristics, may hinder rather than help
investigations by providing producing misleading information.
Recommendation 3: An 'Innocence Project' should be established to investigate
possible miscarriages of justice using DNA.
DNA profiling can be a powerful tool to help establish innocence as well as guilt. DNA
evidence can also be misinterpreted and lead to miscarriages of justice. A project like the US
'Innocence Project' could help increase public confidence that DNA profiling is being used
wisely to improve the criminal justice system in England and Wales. The current system does
not assist in reviews of cases where there is reason to suppose people have been wrongly
convicted.
Recommendation 4: An independent review of whose DNA data should be sampled and
retained is urgently needed. Research on the use of the NDNAD, its effectiveness and
the justification for including innocent people, should be conducted to inform the
debate.
Like others, we are concerned that the legislative changes to date have been introduced too
rapidly and in the absence of any meaningful public debate. We believe that further
GeneWatch UK
8 January 2005deliberation is needed to find out what the public would accept as a reasonable balance
between protecting the right to privacy and protecting citizens from crime. The public should
have a say as to whose data is included on the database and for how long.
There is no data available to evaluate whether crime detection will be improved by including
DNA profiles from people who are arrested and not charged, or by continuing to hold data on
people whose charges are later dropped or who are found to be innocent. GeneWatch UK's
current view is that:
· The personal data and DNA profiles from people whose charges have been
dropped, or who have been acquitted, should be removed from the NDNAD
(unless they were connected with a serious violent or sexual offence). During the
investigation stage, it will have been possible to check whether their profiles match those
from any crime scene on the database. Keeping innocent people on the database
effectively means treating them as criminals. This undermines the principle of 'innocent
until proven guilty' and is open to abuse, particularly in relation to 'political' offences
involving peaceful protest. Making this change would also bring the NDNAD more in line
with record-keeping on the Police National Computer.
· DNA samples should not be taken until a person has been charged, unless
needed to help prove or disprove a suspect's involvement in a specific offence.
The process of charging someone with an offence is more formal than simply making an
arrest. Waiting until a person has been charged reduces the risk that speculative
searching for matches using the database is arbitrary and unfair or that someone could
be 'framed'. This is an important safeguard to prevent the database being used in a
discriminatory way.
· The database should not be expanded to include the whole population. Threats to
privacy and civil rights are more likely to be increased than reduced by proposals to
expand the database; it would not prevent the database from being used in a
discriminatory way and would only serve to increase the potential for state surveillance.
In addition, in the case of a larger database, the probability of false matches and the
resources needed to investigate each match could increase disproportionately to the
number of solved crimes.
· People's personal data and DNA profiles should not be kept indefinitely on the
database (except when they have committed serious violent or sexual crimes).
Keeping records indefinitely raises concerns about civil liberties, particularly when
offences are related to peaceful protest or political dissent. This practice also
undermines the principle of rehabilitation. Records on the Police National Computer are
removed after a certain length of time, depending on the seriousness of the offence.
Records for serious, violent and sexual offences are kept indefinitely, but most other
records are eventually removed. A similar system of restrictions on retention should
apply to the NDNAD.
GeneWatch UK
January 2005 92 Introduction
This report is concerned with the potential threats to privacy and civil liberties posed by the
police use of the National DNA Database (NDNAD) and how these are balanced against the
benefits of using DNA in criminal investigations.
The report provides a brief introduction to the NDNAD and the different ways that the police
use DNA evidence. The details of how DNA samples are obtained, processed and stored are
provided as well as information about how this intelligence is used to trace suspects. The
limitations of the technology, the different sources of error and the need for this evidence to be
used with caution are also discussed in depth.
The report goes on to consider the limits to police powers around the use of the NDNAD as
described in law. Current legislation in England and Wales is compared to that of other
European countries and the United States to explain how the permissive legislation in England
and Wales has enabled the creation of one of the biggest forensic databases in the world.
The future of forensic DNA profiling is also considered, particularly the proposed changes to
the technology and the use of genetic data to predict a suspect's characteristics. The policy
developments likely to impact on police use of the NDNAD and other databases in the future
are considered, including potential threats to privacy and civil liberties.
Finally, the limitations of existing laws and safeguards are discussed and the steps that could
be taken to give greater protection to our rights and freedoms are identified.
GeneWatch UK
10 January 20053 What is the National DNA Database?
The UK National DNA Database (NDNAD) is a police intelligence database that uses DNA to
identify criminal suspects and to find links between different crimes.1 It was set up in April 1995
by the Forensic Science Service (FSS) which has since become a world expert in the use of
forensic DNA technology.
The NDNAD relies on the fact that DNA can be obtained from any sample of human tissue left
2
at the scene of a crime (SOC). This is not just limited to violent crimes where the offender
leaves a sample of blood or semen; the technology is now so sensitive that genetic information
can be extracted from minute samples such as a fingerprint left on a door handle, or the saliva
left on a cup or a cigarette. The extended use of the technology now means that the NDNAD
contains information from a wide range of crimes. Data from every new crime scene is
routinely analysed to see if it matches a known individual on the database or any other SOC
sample. This not only helps to identify suspects, but can also help prove that a person is
innocent.
3
In March 2004, the FSS reported that:
· The NDNAD contained DNA profiles from around 2.1 million individuals and 215,000
crime scenes.
· In a typical month, matches were found linking suspects to 30 murders, 45 rapes and
3,200 motor vehicle, property and drug crimes.
· When information from a new crime scene was added to the database, there was a 40%
chance of an immediate match between the scene of crime (SOC) sample and a known
individual.
· The total value of Government investment in the NDNAD up until that point had been
£182 million, and that a further £61 million was going to be invested in the following year
to support a major expansion of the programme.
The limits on police powers relating to the use of the NDNAD are enshrined in law. On 4 April
2004, these powers were extended. In England and Wales, the police are now allowed to take
samples without consent from anyone who is arrested on suspicion of any recordable offence.
This includes all but the most trivial crimes. The new legislation also allows the police to keep
this information indefinitely, even if the person arrested is never charged. This means the
database is the most extensive in the world and is predicted to expand to include around 5
4
million people (about 10% of the population).
3.1 Using DNA to identify individuals
DNA (deoxyribonucleic acid) is found in every cell in the body. It is passed on from one
generation to the next. Half our DNA is inherited from our mother and half from our father.
Except for identical twins, every single person's DNA is unique. DNA can therefore be used to
try to identify an individual in a similar way to a fingerprint.
However, DNA is very different from other types of forensic data because it has the potential to
5-7
reveal a lot more information about a person. Unlike a fingerprint, DNA can:
· potentially provide some hints about what a person looks like;
· potentially indicate whether a person is at risk of developing an illness in the future or
has a rare genetic condition;
· reveal who a person is related to – if your DNA is held on the NDNAD it could be used to
trace your brothers, sisters, parents or children.
GeneWatch UK
January 2005 11Giving the police access to DNA samples therefore provides them with a lot more information
than a fingerprint does, potentially including personal information about health and
relationships.
3.2 How the police use DNA
If the police obtain a DNA sample from a crime scene there are several ways they can use this
information to find out who it came from:
Þ If there are already one or more known suspects for the crime the police can take DNA
samples from all of them to see if they match the scene of crime (SOC) sample.
Þ If the police suspect that the person who committed the crime lives in a particular village
or works at a particular place, they can ask everyone in that village or workplace to give
a DNA sample in order to try to find a match. These mass screenings can also help to
eliminate a lot of people from police enquiries.
Þ The police can compare the DNA obtained from the crime scene with information on the
NDNAD. If they find a match, the database will give them details about the person they
are trying to find. If they do not find an exact match, they may still be able to identify
relatives of the person they are looking for.
Þ The police can go to court and ask for permission to search other DNA databases for a
match (for example, databases collected for health research), if they can convince the
court that this is in the public interest.
Þ The police can use new techniques to see if the genetic information in the DNA sample
can give some clues about the person they are looking for, such as their hair or skin
colour or their state of health although these techniques currently have serious
limitations. Some researchers claim that it might also be possible in future to predict
certain aspects of behaviour from a person's DNA (such as a tendency to aggression or
addiction) but this is much more controversial.
3.3 Concerns about police use of DNA – our right to privacy
Analysis of DNA can be a powerful tool to help solve crimes. Using DNA evidence wisely can
bring significant benefits to society by helping to convict murderers and rapists. Few people
have problems with the idea of the police comparing the DNA of a suspect with DNA left at the
scene of a serious crime, provided this evidence is used carefully when a case comes to court.
However, concerns arise when tissue samples, genetic information and other types of personal
data are stored on a DNA database. There are fears that this information may be misused in
ways that threaten our individual rights as well as those of our families especially our right to
privacy. Even the police have some concerns about the implications of DNA databases for
their own privacy (see Box T).
Privacy as a right
8
Privacy is a fundamental human right. A UK legal committee has defined it as:
'The right of the individual to be protected against intrusion into his personal life or affairs, or
those of his family, by direct physical means or by publication of information.'
In a free and democratic society, respect for privacy sets an essential limit on the power of
both the state and private organisations to intrude into people's lives. The importance of
protecting privacy is recognised by many international and national treaties including the
9
Universal Declaration of Human Rights and the European Convention on Human Rights,
which has recently been adopted into UK law.8
GeneWatch UK
12 January 2005The collection and storage of personal and genetic information by the police is viewed as a
10
threat to our 'information privacy'. Human genetic data is widely recognised to have special
status because of its ability to reveal private information about a person's health and family
relationships.11 It is therefore argued that all DNA databases, including those held by the
police, deserve special measures of protection. There is also a more general concern that the
other personal data on the NDNAD (and other national databases see Section 8) could
support an increase in population surveillance, taking us closer towards an oppressive 'police
state'.
These fears are not unfounded. Within living memory, both fascist and communist
governments in Europe have used identity papers and other personal records as a means of
oppressing different populations. Rights and freedoms need protection because:
1. they are hard to win and easy to lose;
2. some people may be particularly vulnerable to erosion of their rights, including people
who use mental health services or people from ethnic minorities.
Balancing the interests of the individual and the state
When someone is suspected or convicted of a crime, their rights are restricted in ways that
depend on the seriousness of the offence. Before someone is convicted, there are strict limits
as to how far their rights can be removed. These limits are designed to allow the police to do
their job without giving them so much power that they can act in an arbitrary or unjust way. If
the person is subsequently acquitted, they can then expect to be treated like any other citizen.
This prevents false accusations from seriously damaging people's lives. If the person is
subsequently convicted of a serious offence, they can then expect to be sent to prison and to
lose some of their rights to freedom and privacy.
It is not clear that the police use of DNA databases represents a 'fair and just' restriction of our
right to privacy. The NDNAD contains information about people convicted of a wide range of
offences; people arrested but never charged; people who have been acquitted; and people
who have given their samples voluntarily. They all face the same threats to their rights. The
biggest concern is that the police or Government could use 'the pursuit of criminal justice' to
defend any use of people's personal or genetic information without their knowledge or consent.
There is a danger that the balance of interests between the individual and the state may end
up going too far in one direction (also see Section 9).
GeneWatch UK
January 2005 134 What Information is Stored on the NDNAD?
How is this Information Obtained?
The National DNA Database (NDNAD) contains genetic information in the form of DNA profiles
from both potential suspects and different crime scenes. It also contains more routine
information about people, for example their name and sex. An Arrest Summons Number
allows information on the NDNAD to be linked with information on the Police National
Computer (PNC). The PNC consists of linked databases holding extensive data on criminals,
vehicles and property (see Box O). A barcode reference number also allows information on the
NDNAD to be linked with the corresponding DNA sample, which is kept frozen in storage.
Police forces supply the DNA samples used to derive the DNA profiles on the database, but
individual police officers do not access the NDNAD. The Forensic Science Service (FSS) owns
and manages the database for the Association of Chief Police Officers (ACPO) and supplies
the police with information about matches between DNA profiles.
The details of how this information is obtained and stored are described below.
4.1 DNA profiles
It is useful to know something about the science and technology involved in DNA profiling in
order to understand the strengths and weaknesses of DNA analysis as a tool for crime
investigation.
The technology used to obtain the data for the NDNAD does not examine every single
difference between people's DNA. It is restricted to looking at specific areas known as short
1
tandem repeats (STRs) (see Box A). STRs are places in the DNA where a short section of the
genetic code repeats itself. People have varying numbers of repeats, which is how STRs can
be used to identify individuals.
Ten different STRs are analysed in each DNA sample. Because each STR is made up of two
strands – one from the person's mother and one from their father this analysis produces 20
bits of information.12 The DNA profile then simply consists of a string of numbers indicating the
exact number of repeats at each of the ten STRs plus information about the person's sex.
People may have the same number of repeats at any one STR, but it's the information from all
ten that gives each person their individual profile.
The STRs analysed for DNA profiling are located in non-coding sections of DNA. These are
parts of the DNA that do not contain any instructions for making proteins or any other cell
product. This means they are not part of genes and are not thought to be important in
influencing biological differences such as health or appearance. Therefore, the DNA profiles
themselves are thought to contain very limited amounts of genetic information.
When all the numbers in two complete DNA profiles match exactly, forensic scientists can be
very confident that they come from the same person (except in the case of identical twins).14
The probability of full DNA profiles from two unrelated people matching just by chance is very
low, less than one in a billion.12 However, the chances of a 'false' match are increased (i.e. the
wrong person is identified) if samples are obtained from people who are related; or if the DNA
profile is incomplete. Incomplete profiles are often obtained from SOC samples because the
DNA is often degraded.1 This is one important reason why DNA profiling is not foolproof (see
Section 5.2).
GeneWatch UK
14 January 2005BOX A: Analysing short tandem repeats (STRs) to generate a DNA profile
Diagram of a single STR. We have two copies of each repeat region, one inherited from
our mother and one from our father. The repeated sequence in STRs is typically twoseven
base pairs in length. (A base pair is formed by the connections between two nucleotides
on opposite strands of the DNA. It is equivalent to 'one rung' on the 'DNA ladder'.) The
13
number of times it is repeated usually varies between seven and 30. The short length of
the STRs makes it possible to amplify these sections of DNA using a laboratory technique
2;14
known as PCR. This also makes it possible to obtain DNA profiles from very small
samples and from old samples where the DNA may have degraded.
The end result of DNA profiling is a set of two numbers showing the number of repeats in
each of the ten STRs. In the simplified diagram above this would be '5 6'. A complete
profile would include all 20 STRs and information about gender. A gene called amelogenin
is used for finding out the sex of a sample. Males have X and Y versions; females have
only the X version because they inherit two copies of the X chromosome. So the final DNA
profile of a man would look something like:15
17 17; 14 17; 11 12; 16 17; XY; 13 14; 30 31; 11 23; 14 15; 7 8; 21 23.
The first STR technique, introduced in 1994, looked at only four STRs. The next
development, using SGM (second generation multiplex) profiling, looked at six STRs and
the present-day technology, known as SGM Plus™ , looks at ten STRs. Using a greater
number of STRs increases the discriminatory power of the test, reducing the number of
'false' matches.1
4.2 Obtaining a DNA profile from a tissue sample
The police are responsible for collecting all tissue samples, which they now do routinely as
part of their investigations. Special kits are provided for this purpose: one for obtaining
samples from individuals who have been arrested, known as criminal justice (CJ) samples;
one for taking samples in a mass screening; and one for scene of crime (SOC) samples. There
is also a kit for taking samples from individuals for use as evidence in court; these are known
as case work (CW) samples. Each kit has a unique barcode to allow the sample to be traced
accurately through the system. The process of obtaining a DNA profile from a tissue sample is
described below.
Obtaining a DNA profile from an individual's sample (a CJ sample)16;17
l The process usually begins at a police station when an officer takes a tissue sample from a
newly arrested person. The police first check the Police National Computer (PNC) to see if
the individual's DNA profile is already on the NDNAD a fresh sample is not taken if an
18
existing profile is confirmed. Most often a sample is obtained via a mouth swab a large
cotton wool bud is rubbed inside the suspect's cheek to loosen and collect skin cells.
Alternatively, ten hairs with roots can be removed from the head.
l The sample is then put into a small plastic tube which is sealed. The process is repeated so
that there are two samples for each suspect.
l The police keep their own record of who has given a sample on the PNC database. Each
police record is given an Arrest Summons Number (ASN).
GeneWatch UK
January 2005 15l The sample tubes are labelled with a unique barcode sticker. These are then placed into a
special tamper-proof, clear plastic bag. The bag also contains a card with the same barcode
label and other written details such as the type of alleged crime; the date the sample was
taken; the suspect's name and age; and the name, number and rank of the officer taking the
sample.
l The bag is sealed and the samples sent to an accredited DNA profiling laboratory for
processing. There are currently five organisations that are approved to supply DNA profiles
to the NDNAD: three public laboratories (the FSS and the Strathclyde and Tayside Police
laboratories in Scotland); and two private companies (Cellmark and LGC Limited).
l Staff at the sample reception unit check the integrity of the tamper-proof bag before it is
opened and the card is also checked to ensure all the details have been filled in. A scanner
is used to record the details of each sample onto the computer system.
l One of the two submitted samples from each individual is kept frozen in storage by the
laboratory that carries out the analysis. This 'back up' sample will be kept until the person
would have reached the age 100 if no further information is obtained about them. However,
if it's confirmed that someone has died, the sample will only be kept for one more year
before it is destroyed.
l The other sample is processed to obtain a DNA profile. The DNA is first extracted from the
tissue. It is then amplified in a heating and cooling process that creates millions of copies of
the original DNA – like a biological photocopier. The technical term for this process is the
polymerase chain reaction (PCR).
l The amplified DNA sample is separated out into the 11 specific areas that make up the
profile (ten STRs and the sex marker). A laser detector is used to detect and record the
presence of different DNA fragments distinguished by their size (see Box A).
l The data is analysed by computer which assigns numbers relating to the number of repeats
in each of the STRs. This produces a string of numbers for loading onto the NDNAD.
l The DNA profile is sent by computer to the Forensic Science Service (FSS) to load onto the
NDNAD as part of a new criminal justice (CJ) record. Each CJ record contains the following
information:
- the unique barcode reference number (which provides a link to the stored DNA sample)
- the Arrest Summons Number (which provides a link to the record on the PNC)
- the person's name
- their date of birth
- their ethnic appearance
- their sex
- information about the police force that collected the sample
- information about the laboratory that analysed the sample
- the sample type (blood, semen, saliva etc.)
- the test type
- the DNA profile.
Obtaining a DNA profile from a scene of crime (SOC) sample16
The police collect a variety of evidence from crime scenes in the hope of obtaining a DNA
profile from the perpetrator. This includes samples of blood, semen, hair and urine as well as
19
saliva from cigarette butts, drinking glasses, stamps, envelopes, chewing gum etc. The
sensitivity of the tests has increased to point where even brief contact with a mobile phone,
20
computer keyboard or steering wheel can leave sufficient material to generate a DNA profile.
GeneWatch UK
16 January 2005Scene of crime (SOC) samples go through the same set of processing steps as CJ samples. However, because DNA can be broken down by heat, moisture and sunlight, the DNA in SOC samples is often degraded and the profiles often incomplete. An SOC profile must contain a minimum of eight STRs for it to be loaded onto the NDNAD, although one-off searches can be made with profiles containing only six STRs. Each SOC record contains the following information: - the unique barcode reference number - information about the crime - information about the police force that collected the sample - information about the laboratory that analysed the sample - the sample type (blood, semen, saliva etc.) - the test type - the DNA profile. Who owns the data and the samples? The data and the samples held in storage are the property of the police force that originally collected the sample and sent it to be analysed.21 The FSS owns only the software and IT used to interpret the DNA profiles. It effectively provides a service for the police force, which has to pay for the analysis to be carried out. GeneWatch UK January 2005 17
5 How is the NDNAD Used to Identify Suspects?
How Reliable is DNA Evidence?
5.1 Using the NDNAD to identify suspects
Once new criminal justice (CJ) profiles and DNA profiles from crime scenes (SOC profiles)
have been loaded onto the NDNAD, the computer analyses the results overnight, comparing
16
new profiles with those already on the database. The following morning, 'match' reports are
printed out. There are four types of matches revealing links between:
1. a new SOC profile and the CJ profiles of individuals already on the database. These
matches can identify potential suspects very rapidly.
2. a new CJ profile and the stored SOC profiles from unsolved crimes on the database.
This type of 'speculative search' can identify suspects long after the crime has been
committed.
3. two or more unsolved crimes, when a new SOC profile is found to match that of an old
crime scene, although the identity of the perpetrator is not revealed.
4. a new CJ profile and a CJ profile already on the database, suggesting that the individual
has already been included on the database, under the same or a different name.
Information about matches is sent to the relevant police force purely as intelligence information
(this initial information is not used in court). The police use this data to assist their follow-up
investigations. A match is by no means proof that someone is responsible for a crime. The
19
DNA evidence can show only that someone has probably been present at the crime scene.
The person may well have been there, but not at the time of the incident. It is up to the police
to find all the additional supporting evidence to prove that someone is guilty (see Section 5.3).
Finding suspects via their relatives
Familial searches allow the police to identify a suspect even if their profile is not on the
NDNAD by looking for partial matches between SOC profiles and the profiles of people who
1
might be closely related. The searching process produces a long list of names. Every one of
them has to be contacted by the police to see whether they have a relative who could have
possibly committed the crime.
This approach has been used successfully in a small number of cases (see Box B).1 However,
it is more controversial because it clearly represents a major intrusion into family life – there's
a risk it may uncover family relationships which people do not know about. For example, the
evidence from the 'familial search' might suggest that several people on the database are
related to each other and to the unknown suspect for the crime. But, prior to being contacted
by the police, some of these people may not know who their genetic father is – and may
believe that they are part of an entirely different family.
GeneWatch UK
18 January 2005BOX B: Use of familial searches to find suspects
Case 1: Teenagers Pauline Floyd, Geraldine Hughes and Sandra Newton, all from south
Wales, were raped and strangled in 1973.1 Despite a huge hunt for the murderer at the time,
the cases remained unsolved for more than 29 years. In November 2000, the FSS used DNA
profiling to analyse stains from the clothing of two of the girls but failed to find a match on the
NDNAD. In 2002, DNA profiling was used to test items from the scene of the third girl's murder
and links between the three crimes were revealed for the first time. Again the NDNAD failed to
provide a suspect. Therefore a familial search was adopted to try to find people on the
database who might be related to the perpetrator. The search yielded fewer than 100 names. In
combination with the intelligence already available, this new information led to a local man,
Joseph Kappen, being identified as the prime suspect. But Kappen had already died, so a
proxy DNA profile was created by analysing samples from his family members who had
volunteered to help with the investigation. The results of these tests confirmed that Kappen's
profile was very likely to be a match, which led to his body being exhumed. Subsequent DNA
analysis of his remains showed a perfect match with the SOC samples from all three murders.
Case 2: Craig Harman was the first man to be convicted in Britain following the use of a link
22
between DNA retrieved at the scene of a crime and the DNA profile of a relative. In March
2003, Harman dropped a brick from a footbridge above the M3, which then smashed through
the windscreen of a passing lorry. The driver, Mr Little, suffered a heart attack after the brick hit
him in the chest, and he died. DNA evidence was obtained from fingerprints left on the brick. A
subsequent search of the NDNAD did not produce a match, but did identify a number of people
with similar profiles, including a relative of Mr Harman. It was this crucial piece of evidence that
helped to solve the crime. The match with his relative's DNA helped the police to identify Mr
Harman as a potential suspect. A separate match between his own DNA and the DNA left on
the brick was used as part of the evidence in court.
5.2 DNA profiling is not foolproof
In the case of DNA profiling, as with any other process that relies on people to carry out
actions or make judgements, there is always room for mistakes. These can result in a number
of different problems: the DNA profile may be incorrect although attributed to the right person;
or the profile might be correct but attributed to the wrong person. This can lead to 'false'
matches (or 'false positives') where an innocent person is wrongly accused of a crime.
Alternatively it can lead to 'false negatives' where a criminal escapes detection and is then
2;23-26
able to reoffend.
The possibility of errors therefore undermines the usefulness of the database and the reliability
27 28;29
of DNA evidence. The size of such errors is hard to calculate and is still under dispute.
However, the errors are certainly not negligible and do need to be taken into consideration.
The different sources of error and the steps being taken by the NDNAD to minimise their
impact are described below:
(a) Contamination of samples
DNA samples can become contaminated if they come into contact with DNA from another
person. This contamination results in a DNA profile containing information from an 'outsider'.
This can happen if a sample is mishandled by the police collecting the evidence or the
laboratory staff carrying out the analysis. It is easier to detect in CJ samples because these
should contain only a single profile. It is more difficult in the case of SOC samples, because
these are quite likely to contain DNA from more than one person. Even trace amounts of
outsider DNA can complicate the analysis, so the contamination of samples can lead to
13;24
serious problems. (See section (c) below.)
GeneWatch UK
January 2005 19What measures are taken to reduce the risk of contamination?
Since the whole process is critically dependent on collecting high-quality uncontaminated
13
samples from crime scenes, trained Crime Scene Examiners (CSEs) are employed by police
forces as experts in sample collection. Both the police and the CSEs are required to take
precautionary steps to avoid contamination. These steps include controlling access to the
crime scene; wearing gloves and face masks; and carefully packaging, sealing and labelling
evidence.18
At the processing stage, laboratories use controls and checks to assess the quality of DNA
profiles and to check for contamination. When a problem is picked up, the sample can
sometimes be reanalysed to produce a valid result. Additional tests can also eliminate profiles
12
from laboratory staff and the police. Elimination databases have been set up specifically for
this purpose. These are searched only at the request of a senior officer if they are concerned
that contamination has taken place.
If the source of the contamination is not known or cannot be eliminated, then the result is
considered invalid and the data is not added to the NDNAD. However, it is not clear that all
contaminated samples will be identified.
(b) Poor practice – errors in data handling
These types of error occur when samples are mislabelled during processing or when data is
1
not entered correctly onto the computer. Inevitably, there are always going to be these kinds
of mistakes, but it is important that they do not arise out of sloppy practice. In the USA,
forensic laboratories have been found to routinely mishandle samples and follow incorrect
procedures (see Box C).
BOX C: Poor practice in US police laboratories leads to wrongful convictions
Problems in US police laboratories began in spring 2001,30 when a lab worker realised that
her colleague had not been using the correct procedures to eliminate police DNA profiles
from samples from the scenes of crimes. More than 100 DNA tests were called into
question. The first sample to be retested in Houston, Texas, showed that the DNA profile
31
used to convict a man serving 25 years for rape could not possibly have been his. The
FBI played down the problem, claiming that it was an 'isolated incident' and that it should
not undermine public confidence in its labs. It also announced that it would invest an
additional $1 billion in the US national database to both expand its use and to provide
greater quality assurance.
What measures are taken to reduce the risk of errors in data handling?
In practical terms, new automated systems are being developed to reduce the number of times
14;32
that people have to handle the samples and so reduce the risk of human error. But crucially
the reliability of the results still depends on the people doing the testing. They are therefore
required to go through a regular series of checks. These are overseen by the FSS's Custodian
1;12;21
of the NDNAD and take the form of:
Þ Accreditation: The laboratories that carry out the DNA analysis have to pass a
proficiency test before they are allowed to put profiles on the database. They have to
prove they can consistently produce DNA profiles that are reliable and compatible with
those provided by other suppliers.
Þ Performance monitoring: The laboratories have to prove that they are continuing to
meet the required standards through a programme of internal and external quality audit.
They have to repeat the analysis of 5 to 10% of the samples to show they get consistent
results and demonstrate that they can maintain an error rate below 0.05% (i.e. no more
than one error in every 2,000 samples analysed).
GeneWatch UK
20 January 2005You can also read
